Hi, I'm Sakshi A S

CS Student | NCC Cadet | Problem Solver|Learner

I'm a passionate Computer Science student with a keen interest in web development and in Artificial Intelligence and Deep Learning. I enjoy creating user-centric applications and building meaningful projects that solve real-world problems. My curiosity constantly drives me to explore new technologies and develop impactful tech solutions.

Alongside my academic and technical journey, I proudly serve as a National Cadet Corps (NCC) cadet, which has deeply instilled in me the values of discipline, leadership, resilience, and teamwork. Being part of NCC has shaped me into a well-rounded individual who not only strives for technical excellence but also believes in contributing to society through responsible action and service.

Real-time healthcare assistance system

Care Connect (DTL App) is a real-time healthcare assistance system that enables users to locate the nearest hospital using OpenStreetMap, request an ambulance, and send latitude & longitude to a hospital dashboard. It also features emergency contacts, health profiles, and assistance request options for enhanced emergency support.

Tech: HTML/CSS/JavaScript: For frontend development. Leaflet.js and Routing Machine: For map rendering and route calculation. OpenStreetMap: Provides map data. Firebase: For potential real-time data integration (in progress). LocalStorage: For storing user location, hospital details, and dispatch messages.

GitHub Live Demo

Intrusion detection of networks using GRAPHSAGE and Casual Sampling

An intrusion detection system using Graph Neural Networks (GNNs) and causal sampling to detect malware, IoT attacks, phishing, DDoS attacks, and a global model for all threats. Built with PyTorch Geometric, it processes CICIDS2017 network traffic data into graphs, achieving robust performance.

Tech: Programming Language: Python Frameworks and Libraries: PyTorch: For building and training neural network models. PyTorch Geometric (PyG): For implementing Graph Neural Networks (GNNs), specifically GraphSAGE, to process network traffic as graphs. scikit-learn: For data preprocessing (e.g., LabelEncoder, StandardScaler) and evaluation metrics (e.g., accuracy, F1-score). pandas: For data manipulation and processing CICIDS2017 CSV files into .xlsx format. numpy: For numerical computations and array operations. networkx: For graph construction and analysis (e.g., k-nearest neighbors with cosine similarity). openpyxl: For handling Excel (.xlsx) file I/O.

GitHub

Smart Home Automation System

Developed a cost-effective IoT-based smart home automation system using ESP32/ESP8266 microcontrollers. Integrated ultrasonic and IR sensors for intrusion and motion detection, LDR for light control, and a water pump with level monitoring, all controllable via the Blynk app. Enabled real-time sensor data visualization and remote device management over WiFi, enhancing security, energy efficiency, and convenience.

Tech: ESP32/ESP8266 microcontrollers Arduino IDE Blynk IoT platform WiFi Ultrasonic sensor IR sensor LDR Relay module

GitHub